precompiled sea

classic Classic list List threaded Threaded
16 messages Options
Reply | Threaded
Open this post in threaded view
|

precompiled sea

Mike Baggaley
Hi, I have not used precomp sea before and am struggling to compile the sea
tiles. Using the following process I get a map with no sea:

Used osmfilter to extract just the coastline data
Used splitter on the coastline data
Used mkgmap on the splitter data with a simple style that just processes
coastline and with --generate-sea=multipolygon. I can see some img files
created in my sea folder
Used splitter again on the original data
Used mkgmap again with --precomp-sea and --generate-sea=multipolygon

Am I doing something wrong?

Regards,
Mike

_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Reply | Threaded
Open this post in threaded view
|

Re: precompiled sea

Gerd Petermann
Hi Mike,

not sure what you are trying to do. What do you expect to get with the last command?

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von Mike Baggaley <[hidden email]>
Gesendet: Mittwoch, 29. Januar 2020 19:20
An: 'Development list for mkgmap'
Betreff: [mkgmap-dev] precompiled sea

Hi, I have not used precomp sea before and am struggling to compile the sea
tiles. Using the following process I get a map with no sea:

Used osmfilter to extract just the coastline data
Used splitter on the coastline data
Used mkgmap on the splitter data with a simple style that just processes
coastline and with --generate-sea=multipolygon. I can see some img files
created in my sea folder
Used splitter again on the original data
Used mkgmap again with --precomp-sea and --generate-sea=multipolygon

Am I doing something wrong?

Regards,
Mike

_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Reply | Threaded
Open this post in threaded view
|

Re: precompiled sea

Gerd Petermann
Hi Mike,

if you want to compile your own  sea.zip see
https://wiki.openstreetmap.org/wiki/Mkgmap/help/options#Using_precompiled_sea

If you want to create a map similar to that created with
java -jar mkgmap.jar  --generate-sea=multipolygon --style-file=styles\coastline -c template.args
you have to use something like
java -jar mkgmap.jar --precomp-sea=\osm\sea.zip --style-file=styles\coastline -c template.args

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von Gerd Petermann <[hidden email]>
Gesendet: Mittwoch, 29. Januar 2020 20:22
An: Development list for mkgmap
Betreff: Re: [mkgmap-dev] precompiled sea

Hi Mike,

not sure what you are trying to do. What do you expect to get with the last command?

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von Mike Baggaley <[hidden email]>
Gesendet: Mittwoch, 29. Januar 2020 19:20
An: 'Development list for mkgmap'
Betreff: [mkgmap-dev] precompiled sea

Hi, I have not used precomp sea before and am struggling to compile the sea
tiles. Using the following process I get a map with no sea:

Used osmfilter to extract just the coastline data
Used splitter on the coastline data
Used mkgmap on the splitter data with a simple style that just processes
coastline and with --generate-sea=multipolygon. I can see some img files
created in my sea folder
Used splitter again on the original data
Used mkgmap again with --precomp-sea and --generate-sea=multipolygon

Am I doing something wrong?

Regards,
Mike

_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Reply | Threaded
Open this post in threaded view
|

Re: precompiled sea

Mike Baggaley
In reply to this post by Gerd Petermann
Hi Gerd, up to now I have built my UK map with the sea and land in a single
pass using --generate-sea. However, some tiles are getting the sea and land
inverted, typically where a very small amount of sea is in the tile. I was
trying to see whether if I precompiled the sea, the problem would go away -
I was assuming the tiles of precompiled sea would be bigger because they
contained no other data, so the probability of inversion would be reduced. I
was therefore first trying to generate precompiled sea, then use it to build
my UK map. It may be that my assumption is incorrect and that it won't make
any difference anyway.

Regards,
Mike

-----Original Message-----
From: Gerd Petermann [mailto:[hidden email]]
Sent: 29 January 2020 19:22
To: Development list for mkgmap <[hidden email]>
Subject: Re: [mkgmap-dev] precompiled sea

Hi Mike,

not sure what you are trying to do. What do you expect to get with the last
command?

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von Mike
Baggaley <[hidden email]>
Gesendet: Mittwoch, 29. Januar 2020 19:20
An: 'Development list for mkgmap'
Betreff: [mkgmap-dev] precompiled sea

Hi, I have not used precomp sea before and am struggling to compile the sea
tiles. Using the following process I get a map with no sea:

Used osmfilter to extract just the coastline data
Used splitter on the coastline data
Used mkgmap on the splitter data with a simple style that just processes
coastline and with --generate-sea=multipolygon. I can see some img files
created in my sea folder
Used splitter again on the original data
Used mkgmap again with --precomp-sea and --generate-sea=multipolygon

Am I doing something wrong?

Regards,
Mike

_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev


_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Reply | Threaded
Open this post in threaded view
|

Re: precompiled sea

Mike Baggaley
In reply to this post by Gerd Petermann
Hi Gerd, having now read your link on generating sea yourself, I gather that
mkgmap cannot produce precompiled sea from OSM files, only from ESRI files.
I also wasn't aware that optional code had to be included to use the sea
generator. I could include the optional code in my mkgmap build, but it
looks like using precompiled sea is a non-starter for me if it can't read
OSM data. The original source of my land/sea border was ESRI, but this data
has gone through a filtering process and been converted to OSM format.

I had assumed that the precompiled sea tiles were in img format, but from
opening the precompiled sea zip file from the download page, I see the files
are pbf, suggesting that 'precompiled' is something of a misnomer if they
are just OSM files. I assume it refers to converting from ESRI to OSM, which
I don't need, as it looks like it is doing something pretty similar to what
I already have.

Ah well, back to the drawing board...

Cheers,
Mike

-----Original Message-----
From: Gerd Petermann [mailto:[hidden email]]
Sent: 29 January 2020 20:12
To: Development list for mkgmap <[hidden email]>
Subject: Re: [mkgmap-dev] precompiled sea

Hi Mike,

if you want to compile your own  sea.zip see
https://wiki.openstreetmap.org/wiki/Mkgmap/help/options#Using_precompiled_se
a

If you want to create a map similar to that created with
java -jar mkgmap.jar  --generate-sea=multipolygon
--style-file=styles\coastline -c template.args
you have to use something like
java -jar mkgmap.jar --precomp-sea=\osm\sea.zip
--style-file=styles\coastline -c template.args

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von Gerd
Petermann <[hidden email]>
Gesendet: Mittwoch, 29. Januar 2020 20:22
An: Development list for mkgmap
Betreff: Re: [mkgmap-dev] precompiled sea

Hi Mike,

not sure what you are trying to do. What do you expect to get with the last
command?

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von Mike
Baggaley <[hidden email]>
Gesendet: Mittwoch, 29. Januar 2020 19:20
An: 'Development list for mkgmap'
Betreff: [mkgmap-dev] precompiled sea

Hi, I have not used precomp sea before and am struggling to compile the sea
tiles. Using the following process I get a map with no sea:

Used osmfilter to extract just the coastline data
Used splitter on the coastline data
Used mkgmap on the splitter data with a simple style that just processes
coastline and with --generate-sea=multipolygon. I can see some img files
created in my sea folder
Used splitter again on the original data
Used mkgmap again with --precomp-sea and --generate-sea=multipolygon

Am I doing something wrong?

Regards,
Mike

_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev


_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Reply | Threaded
Open this post in threaded view
|

Re: precompiled sea

Gerd Petermann
Hi Mike,

quite a lot of missunderstandings ;)
1) The data in land-polygons-split-4326.zip is calculated (compiled) from OSM coastline data. AFAIK the same data is used by the renderer of https://www.openstreetmap.org
Find more details  here https://osmdata.openstreetmap.de/data/land-polygons.html
The expected avantage is that the coastline data produced by them is complete and correct (no gaps, no reversed coastline)
2) The data in sea.zip uses the OSM pbf format. It containes the result of processing this (hopefully correct) data using the algorithm that is used with --generate-sea=multipolygon. That's also a reason why it is called pre-compiled.
3) You should load the data of a rather small sea*.pbf file into JOSM to sea what it contains. You'll sea the result of "cutting out holes(islands) of a multipolygon".
So, the file contains only simple ways, no multipolygons.

If you don't want or cannot use precompiled sea data you should try the sub options of --generate-sea.

Gerd
PS: See also my trick regarding testing sea: http://gis.19327.n8.nabble.com/precomp-sea-problem-tp5953261p5953281.html

________________________________________
Von: Mike Baggaley <[hidden email]>
Gesendet: Donnerstag, 30. Januar 2020 00:24
An: 'Gerd Petermann'; 'Development list for mkgmap'
Betreff: RE: [mkgmap-dev] precompiled sea

Hi Gerd, having now read your link on generating sea yourself, I gather that
mkgmap cannot produce precompiled sea from OSM files, only from ESRI files.
I also wasn't aware that optional code had to be included to use the sea
generator. I could include the optional code in my mkgmap build, but it
looks like using precompiled sea is a non-starter for me if it can't read
OSM data. The original source of my land/sea border was ESRI, but this data
has gone through a filtering process and been converted to OSM format.

I had assumed that the precompiled sea tiles were in img format, but from
opening the precompiled sea zip file from the download page, I see the files
are pbf, suggesting that 'precompiled' is something of a misnomer if they
are just OSM files. I assume it refers to converting from ESRI to OSM, which
I don't need, as it looks like it is doing something pretty similar to what
I already have.

Ah well, back to the drawing board...

Cheers,
Mike

-----Original Message-----
From: Gerd Petermann [mailto:[hidden email]]
Sent: 29 January 2020 20:12
To: Development list for mkgmap <[hidden email]>
Subject: Re: [mkgmap-dev] precompiled sea

Hi Mike,

if you want to compile your own  sea.zip see
https://wiki.openstreetmap.org/wiki/Mkgmap/help/options#Using_precompiled_se
a

If you want to create a map similar to that created with
java -jar mkgmap.jar  --generate-sea=multipolygon
--style-file=styles\coastline -c template.args
you have to use something like
java -jar mkgmap.jar --precomp-sea=\osm\sea.zip
--style-file=styles\coastline -c template.args

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von Gerd
Petermann <[hidden email]>
Gesendet: Mittwoch, 29. Januar 2020 20:22
An: Development list for mkgmap
Betreff: Re: [mkgmap-dev] precompiled sea

Hi Mike,

not sure what you are trying to do. What do you expect to get with the last
command?

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von Mike
Baggaley <[hidden email]>
Gesendet: Mittwoch, 29. Januar 2020 19:20
An: 'Development list for mkgmap'
Betreff: [mkgmap-dev] precompiled sea

Hi, I have not used precomp sea before and am struggling to compile the sea
tiles. Using the following process I get a map with no sea:

Used osmfilter to extract just the coastline data
Used splitter on the coastline data
Used mkgmap on the splitter data with a simple style that just processes
coastline and with --generate-sea=multipolygon. I can see some img files
created in my sea folder
Used splitter again on the original data
Used mkgmap again with --precomp-sea and --generate-sea=multipolygon

Am I doing something wrong?

Regards,
Mike

_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev


_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Reply | Threaded
Open this post in threaded view
|

Re: precompiled sea

Ticker Berkin
In reply to this post by Mike Baggaley
Hi

Having just generated full britain-and-ireland with current trunk
(r4432), I'm also now seeing some tiles (5 out of 101) with sea/land
flipped when using option:
  --generate-sea="multipolygon,extend-sea-sectors,close-gaps=350"
but when I process some of the same tiles with r4295 they are OK.

It's not a problem for me at the moment, I've simply replaced above
with
  --precomp-sea=sea-latest.zip

However I prefer to use --generate-sea for various reasons:
- saves downloading sea.zip every now and again.
- island cut-outs match land features exactly, whereas there were
slight differences with sea.zip.
- not had these problems before.
- no noticeable performance problems.

Sometime I'll try and pin down when the change happened. I hadn't
noticed before because the tiles in the small map I use day-to-day
don't show the problem.

Ticker

On Wed, 2020-01-29 at 22:50 +0000, Mike Baggaley wrote:

> Hi Gerd, up to now I have built my UK map with the sea and land in a
> single
> pass using --generate-sea. However, some tiles are getting the sea
> and land
> inverted, typically where a very small amount of sea is in the tile.
> I was
> trying to see whether if I precompiled the sea, the problem would go
> away -
> I was assuming the tiles of precompiled sea would be bigger because
> they
> contained no other data, so the probability of inversion would be
> reduced. I
> was therefore first trying to generate precompiled sea, then use it
> to build
> my UK map. It may be that my assumption is incorrect and that it
> won't make
> any difference anyway.
>
> Regards,
> Mike
>
> -----Original Message-----
> From: Gerd Petermann [mailto:[hidden email]]
> Sent: 29 January 2020 19:22
> To: Development list for mkgmap <[hidden email]>
> Subject: Re: [mkgmap-dev] precompiled sea
>
> Hi Mike,
>
> not sure what you are trying to do. What do you expect to get with
> the last
> command?
>
> Gerd
>
> ________________________________________
> Von: mkgmap-dev <[hidden email]> im Auftrag
> von Mike
> Baggaley <[hidden email]>
> Gesendet: Mittwoch, 29. Januar 2020 19:20
> An: 'Development list for mkgmap'
> Betreff: [mkgmap-dev] precompiled sea
>
> Hi, I have not used precomp sea before and am struggling to compile
> the sea
> tiles. Using the following process I get a map with no sea:
>
> Used osmfilter to extract just the coastline data
> Used splitter on the coastline data
> Used mkgmap on the splitter data with a simple style that just
> processes
> coastline and with --generate-sea=multipolygon. I can see some img
> files
> created in my sea folder
> Used splitter again on the original data
> Used mkgmap again with --precomp-sea and --generate-sea=multipolygon
>
> Am I doing something wrong?
>
> Regards,
> Mike
>
> _______________________________________________
> mkgmap-dev mailing list
> [hidden email]
> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
>
>
> _______________________________________________
> mkgmap-dev mailing list
> [hidden email]
> http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Reply | Threaded
Open this post in threaded view
|

Re: precompiled sea

Ticker Berkin
Hi Gerd, Mike

Considering --generate-sea=... and SeaGenerator.java.

Following changes around r4381, many more tiles have land/sea flipped,
and, looking at the log files, MultiPolygonRelation gives errors for
the sea relation, complaining about inner polygons being in inner
polygons etc.

The code relating to extracting polygons from the coastline doesn't
seem to understand what it is doing regarding the change in behaviour
between land polygons on a sea tile and vice-versa and I have little
confidence in its ability to handle complex coastlines.

So, I've re-written the guts of it to fix these shortcomings. I haven't
touched the "floodblocker" code, which I think should be
decommissioned.

Patch attached. There are a few other aspects that to this code that I
think could be improved.

Ticker

On Thu, 2020-01-30 at 11:35 +0000, Ticker Berkin wrote:

> Hi
>
> Having just generated full britain-and-ireland with current trunk
> (r4432), I'm also now seeing some tiles (5 out of 101) with sea/land
> flipped when using option:
>   --generate-sea="multipolygon,extend-sea-sectors,close-gaps=350"
> but when I process some of the same tiles with r4295 they are OK.
>
> It's not a problem for me at the moment, I've simply replaced above
> with
>   --precomp-sea=sea-latest.zip
>
> However I prefer to use --generate-sea for various reasons:
> - saves downloading sea.zip every now and again.
> - island cut-outs match land features exactly, whereas there were
> slight differences with sea.zip.
> - not had these problems before.
> - no noticeable performance problems.
>
> Sometime I'll try and pin down when the change happened. I hadn't
> noticed before because the tiles in the small map I use day-to-day
> don't show the problem.
>
> Ticker
>
> On Wed, 2020-01-29 at 22:50 +0000, Mike Baggaley wrote:
> > Hi Gerd, up to now I have built my UK map with the sea and land in
> > a
> > single
> > pass using --generate-sea. However, some tiles are getting the sea
> > and land
> > inverted, typically where a very small amount of sea is in the
> > tile.
> > I was
> > trying to see whether if I precompiled the sea, the problem would
> > go
> > away -
> > I was assuming the tiles of precompiled sea would be bigger because
> > they
> > contained no other data, so the probability of inversion would be
> > reduced. I
> > was therefore first trying to generate precompiled sea, then use it
> > to build
> > my UK map. It may be that my assumption is incorrect and that it
> > won't make
> > any difference anyway.
> >
> > Regards,
> > Mike

_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

genSea.patch (35K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: precompiled sea

Gerd Petermann
Hi Ticker,

thanks for the patch. Do you have an example file that shows how 4381 introduced new problems which are now solved?

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von Ticker Berkin <[hidden email]>
Gesendet: Dienstag, 17. März 2020 17:40
An: Development list for mkgmap
Betreff: Re: [mkgmap-dev] precompiled sea

Hi Gerd, Mike

Considering --generate-sea=... and SeaGenerator.java.

Following changes around r4381, many more tiles have land/sea flipped,
and, looking at the log files, MultiPolygonRelation gives errors for
the sea relation, complaining about inner polygons being in inner
polygons etc.

The code relating to extracting polygons from the coastline doesn't
seem to understand what it is doing regarding the change in behaviour
between land polygons on a sea tile and vice-versa and I have little
confidence in its ability to handle complex coastlines.

So, I've re-written the guts of it to fix these shortcomings. I haven't
touched the "floodblocker" code, which I think should be
decommissioned.

Patch attached. There are a few other aspects that to this code that I
think could be improved.

Ticker

On Thu, 2020-01-30 at 11:35 +0000, Ticker Berkin wrote:

> Hi
>
> Having just generated full britain-and-ireland with current trunk
> (r4432), I'm also now seeing some tiles (5 out of 101) with sea/land
> flipped when using option:
>   --generate-sea="multipolygon,extend-sea-sectors,close-gaps=350"
> but when I process some of the same tiles with r4295 they are OK.
>
> It's not a problem for me at the moment, I've simply replaced above
> with
>   --precomp-sea=sea-latest.zip
>
> However I prefer to use --generate-sea for various reasons:
> - saves downloading sea.zip every now and again.
> - island cut-outs match land features exactly, whereas there were
> slight differences with sea.zip.
> - not had these problems before.
> - no noticeable performance problems.
>
> Sometime I'll try and pin down when the change happened. I hadn't
> noticed before because the tiles in the small map I use day-to-day
> don't show the problem.
>
> Ticker
>
> On Wed, 2020-01-29 at 22:50 +0000, Mike Baggaley wrote:
> > Hi Gerd, up to now I have built my UK map with the sea and land in
> > a
> > single
> > pass using --generate-sea. However, some tiles are getting the sea
> > and land
> > inverted, typically where a very small amount of sea is in the
> > tile.
> > I was
> > trying to see whether if I precompiled the sea, the problem would
> > go
> > away -
> > I was assuming the tiles of precompiled sea would be bigger because
> > they
> > contained no other data, so the probability of inversion would be
> > reduced. I
> > was therefore first trying to generate precompiled sea, then use it
> > to build
> > my UK map. It may be that my assumption is incorrect and that it
> > won't make
> > any difference anyway.
> >
> > Regards,
> > Mike
_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Reply | Threaded
Open this post in threaded view
|

Re: precompiled sea

Gerd Petermann
Hi Ticker,

forget it. Found an older extract of Ireland which shows the error with r4382, but not with r4380 or with your patch.
I just want to understand how the changes in 4381 changed this before committing your patch...

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von Gerd Petermann <[hidden email]>
Gesendet: Mittwoch, 18. März 2020 07:41
An: Development list for mkgmap
Betreff: Re: [mkgmap-dev] precompiled sea

Hi Ticker,

thanks for the patch. Do you have an example file that shows how 4381 introduced new problems which are now solved?

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von Ticker Berkin <[hidden email]>
Gesendet: Dienstag, 17. März 2020 17:40
An: Development list for mkgmap
Betreff: Re: [mkgmap-dev] precompiled sea

Hi Gerd, Mike

Considering --generate-sea=... and SeaGenerator.java.

Following changes around r4381, many more tiles have land/sea flipped,
and, looking at the log files, MultiPolygonRelation gives errors for
the sea relation, complaining about inner polygons being in inner
polygons etc.

The code relating to extracting polygons from the coastline doesn't
seem to understand what it is doing regarding the change in behaviour
between land polygons on a sea tile and vice-versa and I have little
confidence in its ability to handle complex coastlines.

So, I've re-written the guts of it to fix these shortcomings. I haven't
touched the "floodblocker" code, which I think should be
decommissioned.

Patch attached. There are a few other aspects that to this code that I
think could be improved.

Ticker

On Thu, 2020-01-30 at 11:35 +0000, Ticker Berkin wrote:

> Hi
>
> Having just generated full britain-and-ireland with current trunk
> (r4432), I'm also now seeing some tiles (5 out of 101) with sea/land
> flipped when using option:
>   --generate-sea="multipolygon,extend-sea-sectors,close-gaps=350"
> but when I process some of the same tiles with r4295 they are OK.
>
> It's not a problem for me at the moment, I've simply replaced above
> with
>   --precomp-sea=sea-latest.zip
>
> However I prefer to use --generate-sea for various reasons:
> - saves downloading sea.zip every now and again.
> - island cut-outs match land features exactly, whereas there were
> slight differences with sea.zip.
> - not had these problems before.
> - no noticeable performance problems.
>
> Sometime I'll try and pin down when the change happened. I hadn't
> noticed before because the tiles in the small map I use day-to-day
> don't show the problem.
>
> Ticker
>
> On Wed, 2020-01-29 at 22:50 +0000, Mike Baggaley wrote:
> > Hi Gerd, up to now I have built my UK map with the sea and land in
> > a
> > single
> > pass using --generate-sea. However, some tiles are getting the sea
> > and land
> > inverted, typically where a very small amount of sea is in the
> > tile.
> > I was
> > trying to see whether if I precompiled the sea, the problem would
> > go
> > away -
> > I was assuming the tiles of precompiled sea would be bigger because
> > they
> > contained no other data, so the probability of inversion would be
> > reduced. I
> > was therefore first trying to generate precompiled sea, then use it
> > to build
> > my UK map. It may be that my assumption is incorrect and that it
> > won't make
> > any difference anyway.
> >
> > Regards,
> > Mike
_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Reply | Threaded
Open this post in threaded view
|

Re: precompiled sea

Gerd Petermann
Hi Ticker,

OK, my results:
The refactoring in r4382 introduced two bugs.
With the partial revert in r4391 both bugs were fixed.
With r4392 one bug was re-introduced.  :(
I've attached a small patch for that.

Now I wonder if your longer patch fixes any other issues?

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von Gerd Petermann <[hidden email]>
Gesendet: Mittwoch, 18. März 2020 08:14
An: Development list for mkgmap
Betreff: Re: [mkgmap-dev] precompiled sea

Hi Ticker,

forget it. Found an older extract of Ireland which shows the error with r4382, but not with r4380 or with your patch.
I just want to understand how the changes in 4381 changed this before committing your patch...

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von Gerd Petermann <[hidden email]>
Gesendet: Mittwoch, 18. März 2020 07:41
An: Development list for mkgmap
Betreff: Re: [mkgmap-dev] precompiled sea

Hi Ticker,

thanks for the patch. Do you have an example file that shows how 4381 introduced new problems which are now solved?

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von Ticker Berkin <[hidden email]>
Gesendet: Dienstag, 17. März 2020 17:40
An: Development list for mkgmap
Betreff: Re: [mkgmap-dev] precompiled sea

Hi Gerd, Mike

Considering --generate-sea=... and SeaGenerator.java.

Following changes around r4381, many more tiles have land/sea flipped,
and, looking at the log files, MultiPolygonRelation gives errors for
the sea relation, complaining about inner polygons being in inner
polygons etc.

The code relating to extracting polygons from the coastline doesn't
seem to understand what it is doing regarding the change in behaviour
between land polygons on a sea tile and vice-versa and I have little
confidence in its ability to handle complex coastlines.

So, I've re-written the guts of it to fix these shortcomings. I haven't
touched the "floodblocker" code, which I think should be
decommissioned.

Patch attached. There are a few other aspects that to this code that I
think could be improved.

Ticker

On Thu, 2020-01-30 at 11:35 +0000, Ticker Berkin wrote:

> Hi
>
> Having just generated full britain-and-ireland with current trunk
> (r4432), I'm also now seeing some tiles (5 out of 101) with sea/land
> flipped when using option:
>   --generate-sea="multipolygon,extend-sea-sectors,close-gaps=350"
> but when I process some of the same tiles with r4295 they are OK.
>
> It's not a problem for me at the moment, I've simply replaced above
> with
>   --precomp-sea=sea-latest.zip
>
> However I prefer to use --generate-sea for various reasons:
> - saves downloading sea.zip every now and again.
> - island cut-outs match land features exactly, whereas there were
> slight differences with sea.zip.
> - not had these problems before.
> - no noticeable performance problems.
>
> Sometime I'll try and pin down when the change happened. I hadn't
> noticed before because the tiles in the small map I use day-to-day
> don't show the problem.
>
> Ticker
>
> On Wed, 2020-01-29 at 22:50 +0000, Mike Baggaley wrote:
> > Hi Gerd, up to now I have built my UK map with the sea and land in
> > a
> > single
> > pass using --generate-sea. However, some tiles are getting the sea
> > and land
> > inverted, typically where a very small amount of sea is in the
> > tile.
> > I was
> > trying to see whether if I precompiled the sea, the problem would
> > go
> > away -
> > I was assuming the tiles of precompiled sea would be bigger because
> > they
> > contained no other data, so the probability of inversion would be
> > reduced. I
> > was therefore first trying to generate precompiled sea, then use it
> > to build
> > my UK map. It may be that my assumption is incorrect and that it
> > won't make
> > any difference anyway.
> >
> > Regards,
> > Mike
_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

fix-r4392.patch (1K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: precompiled sea

Gerd Petermann
Hi Ticker,

I finally understood what was wrong in the addCorners() method, so this patch shows the minimal change to make it work.

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von Gerd Petermann <[hidden email]>
Gesendet: Mittwoch, 18. März 2020 11:05
An: Development list for mkgmap
Betreff: Re: [mkgmap-dev] precompiled sea

Hi Ticker,

OK, my results:
The refactoring in r4382 introduced two bugs.
With the partial revert in r4391 both bugs were fixed.
With r4392 one bug was re-introduced.  :(
I've attached a small patch for that.

Now I wonder if your longer patch fixes any other issues?

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von Gerd Petermann <[hidden email]>
Gesendet: Mittwoch, 18. März 2020 08:14
An: Development list for mkgmap
Betreff: Re: [mkgmap-dev] precompiled sea

Hi Ticker,

forget it. Found an older extract of Ireland which shows the error with r4382, but not with r4380 or with your patch.
I just want to understand how the changes in 4381 changed this before committing your patch...

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von Gerd Petermann <[hidden email]>
Gesendet: Mittwoch, 18. März 2020 07:41
An: Development list for mkgmap
Betreff: Re: [mkgmap-dev] precompiled sea

Hi Ticker,

thanks for the patch. Do you have an example file that shows how 4381 introduced new problems which are now solved?

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von Ticker Berkin <[hidden email]>
Gesendet: Dienstag, 17. März 2020 17:40
An: Development list for mkgmap
Betreff: Re: [mkgmap-dev] precompiled sea

Hi Gerd, Mike

Considering --generate-sea=... and SeaGenerator.java.

Following changes around r4381, many more tiles have land/sea flipped,
and, looking at the log files, MultiPolygonRelation gives errors for
the sea relation, complaining about inner polygons being in inner
polygons etc.

The code relating to extracting polygons from the coastline doesn't
seem to understand what it is doing regarding the change in behaviour
between land polygons on a sea tile and vice-versa and I have little
confidence in its ability to handle complex coastlines.

So, I've re-written the guts of it to fix these shortcomings. I haven't
touched the "floodblocker" code, which I think should be
decommissioned.

Patch attached. There are a few other aspects that to this code that I
think could be improved.

Ticker

On Thu, 2020-01-30 at 11:35 +0000, Ticker Berkin wrote:

> Hi
>
> Having just generated full britain-and-ireland with current trunk
> (r4432), I'm also now seeing some tiles (5 out of 101) with sea/land
> flipped when using option:
>   --generate-sea="multipolygon,extend-sea-sectors,close-gaps=350"
> but when I process some of the same tiles with r4295 they are OK.
>
> It's not a problem for me at the moment, I've simply replaced above
> with
>   --precomp-sea=sea-latest.zip
>
> However I prefer to use --generate-sea for various reasons:
> - saves downloading sea.zip every now and again.
> - island cut-outs match land features exactly, whereas there were
> slight differences with sea.zip.
> - not had these problems before.
> - no noticeable performance problems.
>
> Sometime I'll try and pin down when the change happened. I hadn't
> noticed before because the tiles in the small map I use day-to-day
> don't show the problem.
>
> Ticker
>
> On Wed, 2020-01-29 at 22:50 +0000, Mike Baggaley wrote:
> > Hi Gerd, up to now I have built my UK map with the sea and land in
> > a
> > single
> > pass using --generate-sea. However, some tiles are getting the sea
> > and land
> > inverted, typically where a very small amount of sea is in the
> > tile.
> > I was
> > trying to see whether if I precompiled the sea, the problem would
> > go
> > away -
> > I was assuming the tiles of precompiled sea would be bigger because
> > they
> > contained no other data, so the probability of inversion would be
> > reduced. I
> > was therefore first trying to generate precompiled sea, then use it
> > to build
> > my UK map. It may be that my assumption is incorrect and that it
> > won't make
> > any difference anyway.
> >
> > Regards,
> > Mike
_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev

fix-r4392-v2.patch (744 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: precompiled sea

Ticker Berkin
In reply to this post by Gerd Petermann
Hi Gerd

I was just going through the code and this is what I had so far:

The significant changes are in createInnerWays/createLandPolygons

r4389 createInnerWays extracts land/anticlockwise/islands from the
hitMap. It has some logical mistakes like a 50% chance of putting an
erroneous start/end point on an earlier edge from where the real land
is cut by the tile bounds; this is like a spike, but possibly going
round 3 corners and back!

Then, unless the only shoreline is an enclosed sea
(Caspian/Black/Dead/...?) or there are sea-sectors, it will generated a
sea background and MP.cutout all the land.

Otherwise it generates a land background with this as the MP relation
inner but probably no outers unless there are sea-sectors.

r4392 has similar behaviour and I was just getting to the bit that
you've found where addCorners needed fixing.

My version simplifies and fixes createLandPolygons and addCorners.

It also has other improvements:

If the tile doesn't have any real islands, make it land based and have
any encroaching sea as sea polygons.

If using option sea-sectors:
- Put the triangles on the correct side of the coastline rather than
the 50% chance it had before.
- Don't force the background to land as an "inner" of the triangles
"outer"
- rather, if tile is land, show sea-triangle on the sea-side, if the
background is sea, show a land-triangle on the land-side.

There are a few other minor fixes and tidy-ups

Ticker

On Wed, 2020-03-18 at 10:05 +0000, Gerd Petermann wrote:

> Hi Ticker,
>
> OK, my results:
> The refactoring in r4382 introduced two bugs.
> With the partial revert in r4391 both bugs were fixed.
> With r4392 one bug was re-introduced.  :(
> I've attached a small patch for that.
>
> Now I wonder if your longer patch fixes any other issues?
>
> Gerd

_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Reply | Threaded
Open this post in threaded view
|

Re: precompiled sea

Ticker Berkin
In reply to this post by Gerd Petermann
Hi Gerd

The version in my patch is simpler and works for both
createLandPolygons and createSeaPolygons (both contain a subtle "+= 4")

Ticker

On Wed, 2020-03-18 at 10:26 +0000, Gerd Petermann wrote:
> Hi Ticker,
>
> I finally understood what was wrong in the addCorners() method, so
> this patch shows the minimal change to make it work.
>
> Gerd

_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Reply | Threaded
Open this post in threaded view
|

Re: precompiled sea

David-54
In reply to this post by Mike Baggaley
Hello Ticker and Gerd,

I may introduce some noise in your search. I had a problem with sea generation for Ireland and I wrote about that few weeks ago. Gerd or Ticker suggested to me to update my precompiled sea files. I did it but it did not solve the problem of flooded areas (even with OSM latest data). Then I tried options with generate-sea instead of precompiled sea files, without more success. I decided to wait for few days to see if an update would solve my problem. And it worked. All versions (I did not test later than 4466) after the merge of is_in filter, and few ones before produce, a correct result for Ireland.

David
_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
Reply | Threaded
Open this post in threaded view
|

Re: precompiled sea

Gerd Petermann
Hi David,

yes, sounds like noise unless you did not change anything else.
If you think that a newer version of mkgmap solved the problem please double check with the older mkgmap version so that you can be sure that this is the reason.

Gerd

________________________________________
Von: mkgmap-dev <[hidden email]> im Auftrag von David <[hidden email]>
Gesendet: Mittwoch, 18. März 2020 13:21
An: [hidden email]
Betreff: Re: [mkgmap-dev] precompiled sea

Hello Ticker and Gerd,

I may introduce some noise in your search. I had a problem with sea generation for Ireland and I wrote about that few weeks ago. Gerd or Ticker suggested to me to update my precompiled sea files. I did it but it did not solve the problem of flooded areas (even with OSM latest data). Then I tried options with generate-sea instead of precompiled sea files, without more success. I decided to wait for few days to see if an update would solve my problem. And it worked. All versions (I did not test later than 4466) after the merge of is_in filter, and few ones before produce, a correct result for Ireland.

David
_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev
_______________________________________________
mkgmap-dev mailing list
[hidden email]
http://www.mkgmap.org.uk/mailman/listinfo/mkgmap-dev